visus
/
teensy-audio
Watch 1
Star 0
Fork 0
Code Issues 0 Pull Requests 0 Releases 0 Wiki Activity
Browse Source

Merge pull request #79 from Nantonos/sgtl5000-doc 

Document sgtl5000
dds
Paul Stoffregen 11 years ago
parent
5cc3e4dc4d 27a66cbcb5
commit
13abc7de95
2 changed files with 137 additions and 42 deletions
Split View Show Diff Stats
  1. +7
    -0
      control_sgtl5000.h
  2. +130
    -42
      gui/list.html

+ 7
- 0
control_sgtl5000.h View File

@@ -108,6 +108,13 @@ private:
#define FILTER_PARAEQ 4
#define FILTER_LOSHELF 5
#define FILTER_HISHELF 6
//For frequency adjustment
#define FLAT_FREQUENCY 0
#define PARAMETRIC_EQUALIZER 1
#define TONE_CONTROLS 2
#define GRAPHIC_EQUALIZER 3


void calcBiquad(uint8_t filtertype, float fC, float dB_Gain, float Q, uint32_t quantization_unit, uint32_t fS, int *coef);


+ 130
- 42
gui/list.html View File

@@ -1932,7 +1932,7 @@ double s_freq = .0625;</p>
<p class=desc>Set the headphone volume level. Range is 0 to 1.0, but
0.8 corresponds to the maximum undistorted output for a full scale
signal. Usually 0.5 is a comfortable listening level. The line
level outputs are not changed by this function.
level outputs are <em>not</em> changed by this function.
</p>
<p class=func><span class=keyword>inputSelect</span>(input);</p>
<p class=desc>Select which input to use: AUDIO_INPUT_LINEIN or AUDIO_INPUT_MIC.
@@ -2025,7 +2025,7 @@ double s_freq = .0625;</p>
<p class=func><span class=keyword>adcHighPassFilterFreeze</span>();</p>
<p class=desc>By default, the analog input (either line-level inputs or mic)
is high-pass filtered, to remove any DC component. This function
freezes the filter, so the DC component is still substracted, but
freezes the filter, so the current DC component is still substracted, but
the filter stops tracking any DC or low frequency changes.
</p>
<p class=func><span class=keyword>adcHighPassFilterDisable</span>();</p>
@@ -2035,14 +2035,18 @@ double s_freq = .0625;</p>
<p class=func><span class=keyword>adcHighPassFilterEnable</span>();</p>
<p class=desc>Turn the DC-blocking filter back on, if disabled, or
allows it to resume tracking DC and low frequency changes, if
previously frozen.
previously frozen. This is the default setting.
</p>
<p class=func><span class=keyword>dacVolume</span>(both);</p>
<p class=desc>Normally output volume should be used with volume(), which
changes the analog gain in the headphone amplifier. This function
controls digital attenuation before conversion to analog, which
on the other hand controls digital attenuation before conversion to analog, which
reduces resolution, but allows another fine control of output
signal level. The ranges is 0 to 1.0, with the default at 1.0.
signal level. The ranges is 0 to 1.0, with the default (no digital attenuation)
at 1.0.
</p>
<p class=desc>dacVolume uses zero-crossing detect to avoid clicks, and ramping is handled by
the chip so that a new volume may be set directly in a single call.
</p>
<p class=func><span class=keyword>dacVolume</span>(left, right);</p>
<p class=desc>Adjust the digital output volume separately on left and
@@ -2053,12 +2057,17 @@ double s_freq = .0625;</p>
<h3>Audio Processor</h3>

<p>The optional digital audio processor is capable of implementing
automatic volume control, a
simple equalizer, filtering, bass enhancement and surround sound
is available.
one or more of: automatic volume control, surround sound control,
bass enhancement, and tonal adjustments (either a
simple tone control, or a parametric equalizer, or a graphic equalizer),
in that order.
</p>
<p>These signal processing features are implemented in the SGTL5000 chip,
so they do not consume CPU time on Teensy.
so they do not consume CPU time on Teensy. However, the order of
these processes is fixed in the hardware.
</p>
<p>It is good practice to mute the outputs before enabling or disabling
the Audio Processor, to avoid clicks or thumps.
</p>

<p class=func><span class=keyword>audioPreProcessorEnable</span>();</p>
@@ -2074,59 +2083,138 @@ double s_freq = .0625;</p>
<p class=func><span class=keyword>audioProcessorDisable</span>();</p>
<p class=desc>Disable the audio processor.
</p>
<p class=func><span class=keyword>eqFilterCount</span>(n);</p>
<p class=desc>blah blah blah blah
<p class=func><span class=keyword>autoVolumeControl</span>(maxGain, response, hardLimit, threshold, attack, decay);</p>
<p class=desc>Configures the auto volume control, which is implemented as a compressor/expander
or hard limiter. <em>maxGain</em> is the maximum gain that can be applied for expanding, and
can take one of three values: 0 (0dB), 1 (6.0dB) and 2 (12dB). Values greater than 2 are treated
as 2. <em>response</em> controls the integration time for the compressor and can take
four values: 0 (0ms), 1 (25ms), 2 (50ms) or 3 (100ms). Larger values average the volume
over a longer time, allowing short-term peaks through.
</p>
<p class=desc>If <em>hardLimit</em> is 0, a 'soft
knee' compressor is used to progressively compress louder values which are near to or above the
threashold (the louder they are, the greater the compression). If it is 1, a hard compressor
is used (all values above the threashold are the same loudness). The <em>threashold</em> is specified
as a float in the range 0dBFS to -96dBFS, where -18dBFS is a typical value.
<em>attack</em> is a float controlling the rate of decrease in gain when the signal is over
threashold, in dB/s. <em>decay</em> controls how fast gain is restored once the level
drops below threashold, again in dB/s. It is typically set to a longer value than attack.
</p>
<p class=func><span class=keyword>eqSelect</span>(n);</p>
<p class=desc>blah blah blah blah
<p class=func><span class=keyword>autoVolumeEnable</span>();</p>
<p class=desc>Enables auto volume control, using the previously specified settings.
</p>
<p class=func><span class=keyword>eqBand</span>(bandNum, n);</p>
<p class=desc>blah blah blah blah
<p class=func><span class=keyword>autoVolumeDisable</span>();</p>
<p class=desc>Disables auto volume control.
</p>
<p class=func><span class=keyword>eqBands</span>(bass, mid_bass, midrange, mid_treble, treble);</p>
<p class=desc>blah blah blah blah

<p class=func><span class=keyword>surroundSoundEnable</span>();</p>
<p class=desc>Enable virtual surround processing, to give a broader and
deeper stereo image (even with mono input).
</p>
<p class=func><span class=keyword>eqBands</span>(bass, treble);</p>
<p class=desc>blah blah blah blah
<p class=func><span class=keyword>surroundSoundDisable</span>();</p>
<p class=desc>Disable virtual surround processing. Before disabling, ramp up
the width to maximum to avoid pops.
</p>
<p class=func><span class=keyword>eqFilter</span>(filterNum, filterParameters);</p>
<p class=desc>blah blah blah blah
<p class=func><span class=keyword>surroundSound</span>(width);</p>
<p class=desc>Configures virtual surround width from 0 (mono) to 7 (widest).
</p>
<p class=func><span class=keyword>autoVolumeControl</span>(maxGain, response, hardLimit, threshold, attack, decay);</p>
<p class=desc>blah blah blah blah
<p class=func><span class=keyword>surroundSound</span>(width, select);</p>
<p class=desc>Configures virtual surround width from 0 (mono) to 7 (widest).
<em>select</em> may be set to 1 (disable), 2 (mono input) or 3 (stereo input).
</p>
<p class=func><span class=keyword>autoVolumeEnable</span>();</p>
<p class=desc>blah blah blah blah
<p class=func><span class=keyword>enhanceBassEnable</span>();</p>
<p class=desc>Enable bass enhancement. A mono, low-pass filtered copy of
the original stereo signal has bass levels boosted and is then mixed back into
the stereo signal, which is then optionally high pass filtered (to remove
inaudible subsonic frequencies).
</p>
<p class=func><span class=keyword>autoVolumeDisable</span>();</p>
<p class=desc>blah blah blah blah
<p class=func><span class=keyword>enhanceBassDisable</span>();</p>
<p class=desc>Disable bass enhancement. Before disabling, ramp down the bass
enhancement level to zero.
</p>
<p class=func><span class=keyword>enhanceBass</span>(lr_lev, bass_lev);</p>
<p class=desc>blah blah blah blah
<p class=desc>Configures the bass enhancement by setting the levels of the
original stereo signal and the bass-enhanced mono level which will be mixed together.
There is no high-pass filter.
</p>
<p class=desc>When changing bass level, call this function repeatedly to ramp up or down the bass in
steps of 0.5dB, to avoid pops.
</p>
<p class=func><span class=keyword>enhanceBass</span>(lr_lev, bass_lev, hpf_bypass, cutoff);</p>
<p class=desc>blah blah blah blah
<p class=desc>Configures the bass enhancement by setting the levels of the
original stereo signal and the bass-enhanced mono level which will be mixed together.
The high-pass filter may be enabled (0) or bypassed (1). The cutoff frequency is specified
as follows:
</p>
<pre class="desc">
value frequency
0 80Hz
1 100Hz
2 125Hz
3 150Hz
4 175Hz
5 200Hz
6 225Hz
</pre>
<p class=desc>When changing bass level, call this function repeatedly to ramp up or down the bass in
steps of 0.5dB, to avoid pops.
</p>
<p class=func><span class=keyword>enhanceBassEnable</span>();</p>
<p class=desc>blah blah blah blah
<p class=func><span class=keyword>eqSelect</span>(n);</p>
<p class=desc>Selects the type of frequency control, where <em>n</em> is
one of</p>
<p class=desc><b>FLAT_FREQUENCY (0)</b><br>
Equalizers and tone controls disabled, flat frequency response.</p>
<p class=desc><b>PARAMETRIC_EQUALIZER (1)</b><br>
Enables the 7-band parametric equalizer, thus disabling the
tone controls and graphic equalizer.</p>
<p class=desc><b>TONE_CONTROLS (2)</b><br>
Enables bass and treble tone controls, disabling the parametric
equalization and graphic equalizer.</p>
<p class=desc><b>GRAPHIC_EQUALIZER (3)</b><br>
Enables the five-band graphic equalizer, disabling the parametric
equalization and tone controls.</p>

<p class=func><span class=keyword>eqBands</span>(bass, treble);</p>
<p class=desc>Configures bass and treble tone controls, which are
implemented as one second order low pass filter (bass) in parallel with
one second order high pass filter (treble).
</p>
<p class=func><span class=keyword>enhanceBassDisable</span>();</p>
<p class=desc>blah blah blah blah
<p class=desc>When changing bass or treble level, call this function repeatedly to ramp
up or down the level in steps of 0.5dB, to avoid pops.
</p>
<p class=func><span class=keyword>surroundSound</span>(width);</p>
<p class=desc>blah blah blah blah
<p class=func><span class=keyword>eqBands</span>(bass, mid_bass, midrange, mid_treble, treble);</p>
<p class=desc>Configures the graphic equalizer. It is implemented by five parallel,
second order biquad filters with fixed frequencies of 115Hz, 330Hz, 990Hz, 3kHz,
and 9.9kHz. Each band has a range of adjustment from 100.0 (+12dB) to -100.0 (-11.75dB).
</p>
<p class=func><span class=keyword>surroundSound</span>(width, select);</p>
<p class=desc>blah blah blah blah
<p class=func><span class=keyword>eqBand</span>(bandNum, n);</p>
<p class=desc>Configures the gain or cut on one band in the graphic equalizer.
<em>bandnum</em> can range from 1 to 5; <em>n</em> is a float in the range 100.0 to -100.0.
</p>
<p class=func><span class=keyword>surroundSoundEnable</span>();</p>
<p class=desc>blah blah blah blah
<p class=desc>When changing a band, call this function repeatedly to ramp up the gain in steps of 0.5dB,
to avoid pops.
</p>
<p class=func><span class=keyword>surroundSoundDisable</span>();</p>
<p class=desc>blah blah blah blah
<p class=func><span class=keyword>eqFilter</span>(filterNum, filterParameters);</p>
<p class=desc>Configurs the parametric equalizer. The number of filters (1 to 7)
is specified along with a pointer to an array of filter coefficients.
The parametric equalizer is implemented using 7 cascaded, second order bi-quad
filters whose frequencies, gain, and Q may be freely configured, but each filter
can only be specified as a set of filter coefficients.
</p>
<p class=func><span class=keyword>eqFilterCount</span>(n);</p>
<p class=desc>Enables zero or more of the already enabled parametric filters.
</p>
<h3>Notes</h3>
<p>TODO: document the many audio processor features &amp; functions!
<p>TODO: add example with rock/classical/speech presets, where rock uses bass boost
and surround enhancement while speech uses bandpass filtering and auto volume control
compression.
</p>
<p>TODO: add example with two analogRead pots for bass and treble to demonstrate ramping.
</p>
</script>
<script type="text/x-red" data-template-name="AudioControlSGTL5000">

Write Preview
Loading…
Cancel
Save